Deep sleep shutdown the phone

[lacriz]

For Mobian using its default bootloader off eMMC I have no problems.

Hi dukia,

thank you a lot for your response. So just that I am clear before I switch (back) to mobian (I still have it on an sd card, so I will definitively try it - I just want it to be documented here for everyone's sake):

In Mobian, when you do 'systemctl suspend' (or somesuch, I put that in a 1-line-script) or just wait your X configured minutes for deep sleep to kick-in:
1) the phone goes 'off'
2) when you press the power button -shortly-,
3) the phone wakes up again (without showing a boot screen?)
4) you have to login again (I guess - would be fine for me although very aggravating imho)
5) all previously running apps are still running
6) those apps are still useable as-is
7) those apps are still in the exact same state as before step 1)
8) icying on the cake: the uptime is > 0

Is that how it behaves when using Mobian?


Thank you in advance


   lacriz

[bcnaz]

That is pretty good description of the 'current' Bookworm Nightly fresh install to sd card running on PMOS Convergent phone, just '"Outta the Box"' -- so to speak.

(though at the freshest release there is always a risk of something breaking, but the Mobian Team usually has it fixed by the next release.)

[lacriz]

I sadly have to report that deep sleep is not working on my 3GB/32GB beta convergence PinePhone. It never has been working, on neither of these distros:

Manjaro + Plasma Mobile (pre-installed)
arch + phosh (from eMMC)
mobian-pinephone-phosh-20210517.img ('stable' image from sd card)
mobian-pinephone-phosh-20211018.img ('nightly image from sd card, flashed that just now)

I can only assume it is a crust issue or somehow dependant on the hardware itself.

I would be kind of interested in advice on how to proceed from here.

However, I somewhat gave up on this issue after spending way too many hours into that topic. I can live with auto-starting firefox, signal, email, calendar and weather app. I set this up using the gnome 'tweaks' app which frees me from having to use the Terminal for such a basic task - which of course no iOS or Android phone even offers

[bcnaz]

Over-all, deep sleep appears to work for me, but there are some faults, * if the charge is below 15% when it is in deep sleep, it will crash.

It froze a few times a few weeks back, but then the problem disappeared after some updates.

Then again once the night before last it froze, (again the charge was at 15%) after a hard reset it was running again.

This is on the phone I do use as my main phone, the OS was originally flashed about August 15th (I think), with updates since.

>> My test phone that has the most recent nightly release has not had the freezing problem (Yet)

Both are Mobian Bookworm

[lacriz]

Over-all,  deep sleep appears to work for me

Hi bcnaz,

thank you a lot for posting your insights.
Would you mind telling us on which exact PinePhone model deep sleep works for you?

Is it the beta edition with 3 GB RAM? Or an older model?


Thanks for clarifying

   lacriz

[SocialNetworkingWasAMistake]

Are you sure your power button is actually working? I know mine died

[lacriz]

Are you sure your power button is actually working? I know mine died

I am not sure to whom your question is directed. Sorry for assuming it is me: I -think- my power button works:
1) when I press it shortly, the display goes off / on
2) when I press it for about 20 seconds, a restart is forced

To actually trigger the deep sleep functionality, I know only of 2 ways so far:
a) wait x Minutes (depends on your settings, makes no sense if auto-deep sleep is disabled there)
b) sudo systemctl suspend

[bcnaz]

Over-all,  deep sleep appears to work for me

Hi bcnaz,

thank you a lot for posting your insights.
Would you mind telling us on which exact PinePhone model deep sleep works for you?

Is it the beta edition with 3 GB RAM? Or an older model?


Thanks for clarifying

   lacriz

The daily driver is a rev: 1.2a 3g ram,   while the test phone is the rev: 1.2b 3g ram

the 1.2a is the one that I noticed the problem on.

[lacriz]

Thank you bcnaz, now I can rule out that the 3GB-models in general are to blame.
FWIW, I have a 1.2b revision (beta edition).

I followed kqlnut's advice and ran journalctl after a 'wake-from-sleep' boot sequence:

Code:

journalctl --since "2min ago"

Here are the -hopefully- relevant messages; does any of that sound familiar to anyone? [How] can I fix this?

Code:

Okt 22 12:41:25 danctnix systemd[1]: Activating swap /dev/zram0...
Okt 22 12:41:25 danctnix systemd[1]: Condition check resulted in Kernel Trace File System being skipped.
Okt 22 12:41:25 danctnix systemd[1]: Condition check resulted in Entropy Daemon based on the HAVEGE algorithm being sk>
Okt 22 12:41:25 danctnix systemd[1]: Condition check resulted in Set Up Additional Binary Formats being skipped.
Okt 22 12:41:25 danctnix systemd[1]: Condition check resulted in First Boot Wizard being skipped.
Okt 22 12:41:25 danctnix systemd[1]: Condition check resulted in First Boot Complete being skipped.
Okt 22 12:41:25 danctnix systemd[1]: Condition check resulted in File System Check on Root Device being skipped.
Okt 22 12:41:25 danctnix systemd[1]: Condition check resulted in Rebuild Hardware Database being skipped.
Okt 22 12:41:25 danctnix systemd[1]: Condition check resulted in Repartition Root Disk being skipped.
Okt 22 12:41:25 danctnix systemd[1]: Condition check resulted in Create System Users being skipped.
Okt 22 12:41:25 danctnix swapon[2913]: swapon: /dev/zram0: read swap header failed
Okt 22 12:41:25 danctnix systemd[1]: dev-zram0.swap: Swap process exited, code=exited, status=255/EXCEPTION
Okt 22 12:41:25 danctnix systemd[1]: dev-zram0.swap: Failed with result 'exit-code'.
Okt 22 12:41:25 danctnix systemd[1]: Failed to activate swap /dev/zram0.
Okt 22 12:41:25 danctnix systemd[1]: Dependency failed for Swaps.
Okt 22 12:41:25 danctnix systemd[1]: swap.target: Job swap.target/start failed with result 'dependency'.
Okt 22 12:41:25 danctnix systemd[1]: Mounting Temporary Directory /tmp...


Okt 22 12:41:36 danctnix dbus-daemon[2943]: [system] Activating via systemd: service name='org.freedesktop.UPower' uni>
Okt 22 12:41:36 danctnix systemd[1]: Starting Daemon for power management...
Okt 22 12:41:37 danctnix upowerd[3226]: did not recognise USB path /sys/devices/platform/soc/1f03400.rsb/sunxi-rsb-3a3>
Okt 22 12:41:37 danctnix dbus-daemon[2943]: [system] Successfully activated service 'org.freedesktop.UPower'
Okt 22 12:41:37 danctnix systemd[1]: Started Daemon for power management.

Okt 22 12:41:40 danctnix gnome-session[3189]: gnome-session-binary[3189]: WARNING: Could not retrieve current screensa>
Okt 22 12:41:40 danctnix gnome-session-binary[3189]: WARNING: Could not retrieve current screensaver active state: Tim>
Okt 22 12:41:40 danctnix upowerd[3226]: treating change event as add on /sys/devices/platform/soc/1c1b000.usb/usb2/2-1

[kqlnut]

Code:

journalctl --since "2min ago"

Here are the -hopefully- relevant messages; does any of that sound familiar to anyone? [How] can I fix this?

Without knowing too much of it, it looks like a regular boot sequence. So it would be more helpful to look at the log from the end of the last boot, i. e. when the phone suspended and crashed.